In this study, patients with advanced adenocarcinoma that reacts with NR-LU-10 monoclonal antibody will be treated with radioimmunotherapy using a pretargeting approach. Using this method, each patient will receive a total of three infusions. The first infusion will consist of NR-LU-10 monoclonal antibody/Streptavidin, followed by a clearing agent (Calactose-Biotin-Human Serum Albumin) to remove circulating unbound antibody, and lastly, 111In and 90Y-DOTA-Biotin. Patients will then be imaged in the anterior and posterior projections in a whole body mode immediately following injection and at three hours and one, two, and five days following treatment. Dosed to tumor and normal tissue will be calculated from data obtained from the immunoscintigraphy studies as well as tumor and bone marrow biopsies in a subset of patients. A minimum of three patients will be treated at each dose level until the maximal tolerated dose is established. Tumor size measurements will be obtained whenever possible. Clinical efficacy will be assessed using a variety of imaging studies appropriate for the particular patients known/monitored sites of disease, as well as by serum tumor marker measurements when indicated. In addition, the incidence and timing of human anti-immunoconjugate antibody responses will be determined by ELISA assays. Pharmacokinetic profiles will be determined by measuring the levels of immunoconjugate in serum either by ELISA or RIA assays. The primary objectives of the study are to establish the MTD of 90Y-Biotin-Dota when administered as part of a pretargeting approach to evaluate the safety of the pretargeting approach. To date 26 patients have been treated on this study at dose levels ranging from 50mCi/m2 to 140mCi/m2. At the higher doses there have been site dependent reductions in tumor size in some of the patients, two PRS but no CRS. The only significant toxicities noted to date are reversible myelosuppression and gastrointestinal side effects. Diarrhea was the dose limiting toxicity at 120 and 140 mCi/m2. Three additional patients with no prior pelvic of abdominal radiotherapy will be treated at the 120mCi/m2 dose level and then one study will be closed to further patient accrual. The Amgen KGF study includes adults with unreactable advanced head and neck cancers. Patients receive concomitant twice-daily irradiation and cyclic cisplatin-5FU chemotherapy. The randomization( and the need for the GCRC) involves KGF (Keratinocyte Growth Factor) versus placebo, given during the radiotherapy course. If the KGF works as expected, those will be less mucositis in this arm.